[gentoo-portage-dev] Re: [PATCH] pkg_use_display: show masked/forced state of USE_EXPAND flags (bug 490562)

[Zac Medico]

On Wed, Mar 22, 2017 at 9:32 AM, Zac Medico  wrote:
> Fix pkg_use_display to test if the prefixed flag is in use.force or
> use.mask, rather than the unprefixed flag.
>
> X-Gentoo-bug: 490562
> X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=490562
> ---
>  pym/_emerge/UseFlagDisplay.py | 19 +--
>  1 file changed, 13 insertions(+), 6 deletions(-)

Bump.
-- 
Thanks,
Zac

Re: [gentoo-portage-dev] [PATCH] pkg_use_display: show masked/forced state of USE_EXPAND flags (bug 490562)

[Brian Dolbec]

On Wed, 22 Mar 2017 09:32:36 -0700
Zac Medico  wrote:

> Fix pkg_use_display to test if the prefixed flag is in use.force or
> use.mask, rather than the unprefixed flag.
> 
> X-Gentoo-bug: 490562
> X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=490562
> ---
>  pym/_emerge/UseFlagDisplay.py | 19 +--
>  1 file changed, 13 insertions(+), 6 deletions(-)
> 
> diff --git a/pym/_emerge/UseFlagDisplay.py
> b/pym/_emerge/UseFlagDisplay.py index f460474..12820e9 100644
> --- a/pym/_emerge/UseFlagDisplay.py
> +++ b/pym/_emerge/UseFlagDisplay.py
> @@ -3,6 +3,7 @@
>  
>  from __future__ import unicode_literals
>  
> +import collections
>  from itertools import chain
>  import sys
>  
> @@ -60,6 +61,10 @@ class UseFlagDisplay(object):
>   sort_separated = cmp_sort_key(_cmp_separated)
>   del _cmp_separated
>  
> +
> +_flag_info = collections.namedtuple('_flag_info', ('flag',
> 'display')) +
> +
>  def pkg_use_display(pkg, opts, modified_use=None):
>   settings = pkg.root_config.settings
>   use_expand = pkg.use.expand
> @@ -81,27 +86,29 @@ def pkg_use_display(pkg, opts, modified_use=None):
>   if f.startswith(flag_prefix):
>   use_expand_flags.add(f)
>   use_enabled.setdefault(
> - varname.upper(),
> []).append(f[len(flag_prefix):])
> + varname.upper(), []).append(
> + _flag_info(f,
> f[len(flag_prefix):])) 
>   for f in pkg.iuse.all:
>   if f.startswith(flag_prefix):
>   use_expand_flags.add(f)
>   if f not in use:
>   use_disabled.setdefault(
> - varname.upper(),
> []).append(f[len(flag_prefix):])
> + varname.upper(),
> []).append(
> +
> _flag_info(f, f[len(flag_prefix):])) 
>   var_order = set(use_enabled)
>   var_order.update(use_disabled)
>   var_order = sorted(var_order)
>   var_order.insert(0, 'USE')
>   use.difference_update(use_expand_flags)
> - use_enabled['USE'] = list(use)
> + use_enabled['USE'] = list(_flag_info(f, f) for f in use)
>   use_disabled['USE'] = []
>  
>   for f in pkg.iuse.all:
>   if f not in use and \
>   f not in use_expand_flags:
> - use_disabled['USE'].append(f)
> + use_disabled['USE'].append(_flag_info(f, f))
>  
>   flag_displays = []
>   for varname in var_order:
> @@ -109,9 +116,9 @@ def pkg_use_display(pkg, opts, modified_use=None):
>   continue
>   flags = []
>   for f in use_enabled.get(varname, []):
> - flags.append(UseFlagDisplay(f, True, f in
> forced_flags))
> + flags.append(UseFlagDisplay(f.display, True,
> f.flag in forced_flags)) for f in use_disabled.get(varname, []):
> - flags.append(UseFlagDisplay(f, False, f in
> forced_flags))
> + flags.append(UseFlagDisplay(f.display,
> False, f.flag in forced_flags)) if alphabetical_use:
>   flags.sort(key=UseFlagDisplay.sort_combined)
>   else:

Sorry, Zac, I somehow missed this one...  too busy with my head burried
in twisted test fixes and new deps...

Looks fine :)

-- 
Brian Dolbec 

Re: [gentoo-portage-dev] [PATCH] pkg_use_display: show masked/forced state of USE_EXPAND flags (bug 490562)

[Zac Medico]

On Fri, Mar 31, 2017 at 1:05 PM, Brian Dolbec  wrote:
> On Wed, 22 Mar 2017 09:32:36 -0700
> Zac Medico  wrote:
>
>> Fix pkg_use_display to test if the prefixed flag is in use.force or
>> use.mask, rather than the unprefixed flag.
>>
>> X-Gentoo-bug: 490562
>> X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=490562
>
> Sorry, Zac, I somehow missed this one...  too busy with my head burried
> in twisted test fixes and new deps...
>
> Looks fine :)

Thanks, pushed:

https://gitweb.gentoo.org/proj/portage.git/commit/?id=09ee3124256c2bb0785e246533b62d310d54082e
-- 
Thanks,
Zac

[gentoo-portage-dev] [PATCH] depgraph: trigger slot operator rebuilds via _complete_graph (bug 614390)

[Zac Medico]

Fix _complete_graph to trigger rebuilds of parent packages when they
pull in installed packages that had already been scheduled for rebuild
by the previous calculation.

X-Gentoo-bug: 614390
X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=614390
---
 pym/_emerge/depgraph.py|  15 +++
 .../resolver/test_slot_operator_complete_graph.py  | 141 +
 2 files changed, 156 insertions(+)
 create mode 100644 
pym/portage/tests/resolver/test_slot_operator_complete_graph.py

diff --git a/pym/_emerge/depgraph.py b/pym/_emerge/depgraph.py
index 04e724d..8a614c4 100644
--- a/pym/_emerge/depgraph.py
+++ b/pym/_emerge/depgraph.py
@@ -6649,6 +6649,21 @@ class depgraph(object):
# will be appropriately reported as a slot 
collision
# (possibly solvable via backtracking).
pkg = matches[-1] # highest match
+
+   if (self._dynamic_config._allow_backtracking and
+   not self._want_installed_pkg(pkg) and 
(dep.atom.soname or (
+   dep.atom.package and 
dep.atom.slot_operator_built))):
+   # If pkg was already scheduled for 
rebuild by the previous
+   # calculation, then pulling in the 
installed instance will
+   # trigger a slot conflict that may go 
unsolved. Therefore,
+   # trigger a rebuild of the parent if 
appropriate.
+   dep.child = pkg
+   new_dep = 
self._slot_operator_update_probe(dep)
+   if new_dep is not None:
+   
self._slot_operator_update_backtrack(
+   dep, new_dep=new_dep)
+   continue
+
if not self._add_pkg(pkg, dep):
return 0
if not 
self._create_graph(allow_unsatisfied=True):
diff --git a/pym/portage/tests/resolver/test_slot_operator_complete_graph.py 
b/pym/portage/tests/resolver/test_slot_operator_complete_graph.py
new file mode 100644
index 000..1d59bce
--- /dev/null
+++ b/pym/portage/tests/resolver/test_slot_operator_complete_graph.py
@@ -0,0 +1,141 @@
+# Copyright 2017 Gentoo Foundation
+# Distributed under the terms of the GNU General Public License v2
+
+from portage.tests import TestCase
+from portage.tests.resolver.ResolverPlayground import (
+   ResolverPlayground,
+   ResolverPlaygroundTestCase,
+)
+
+class SlotOperatorCompleteGraphTestCase(TestCase):
+
+   def testSlotOperatorCompleteGraph(self):
+
+   ebuilds = {
+   "app-misc/meta-pkg-2" : {
+   "EAPI": "6",
+   "DEPEND": "=app-misc/B-2 =app-misc/C-1  
=app-misc/D-1 =dev-libs/foo-2",
+   "RDEPEND": "=app-misc/B-2 =app-misc/C-1 
=app-misc/D-1 =dev-libs/foo-2",
+   },
+
+   "app-misc/meta-pkg-1" : {
+   "EAPI": "6",
+   "DEPEND": "=app-misc/B-1 =app-misc/C-1  
=app-misc/D-1 =dev-libs/foo-1",
+   "RDEPEND": "=app-misc/B-1 =app-misc/C-1 
=app-misc/D-1 =dev-libs/foo-1",
+   },
+
+   "app-misc/B-1" : {
+   "EAPI": "6",
+   "DEPEND": "dev-libs/foo:=",
+   "RDEPEND": "dev-libs/foo:=",
+   },
+
+   "app-misc/B-2" : {
+   "EAPI": "6",
+   "DEPEND": "dev-libs/foo:=",
+   "RDEPEND": "dev-libs/foo:=",
+   },
+
+   "app-misc/C-1" : {
+   "EAPI": "6",
+   "DEPEND": "dev-libs/foo:= app-misc/B",
+   "RDEPEND": "dev-libs/foo:= app-misc/B",
+   },
+
+   "app-misc/C-2" : {
+   "EAPI": "6",
+   "DEPEND": "dev-libs/foo:= app-misc/B",
+   "RDEPEND": "dev-libs/foo:= app-misc/B",
+   },
+
+   "app-misc/D-1" : {
+   "EAPI": "6",
+   "DEPEND": "dev-libs/foo:=",
+   "RDEPEND": "dev-libs/foo:=",
+   },
+
+   "app-misc/D-2" : {
+   "EAPI": "6",
+   "DEPEND": "dev-libs/foo:=",
+ 

Re: [gentoo-portage-dev] [PATCH] depgraph: trigger slot operator rebuilds via _complete_graph (bug 614390)

[Brian Dolbec]

On Fri, 31 Mar 2017 21:11:41 -0700
Zac Medico  wrote:

> Fix _complete_graph to trigger rebuilds of parent packages when they
> pull in installed packages that had already been scheduled for rebuild
> by the previous calculation.
> 
> X-Gentoo-bug: 614390
> X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=614390
> ---
>  pym/_emerge/depgraph.py|  15 +++
>  .../resolver/test_slot_operator_complete_graph.py  | 141
> + 2 files changed, 156 insertions(+)
>  create mode 100644
> pym/portage/tests/resolver/test_slot_operator_complete_graph.py
> 
> diff --git a/pym/_emerge/depgraph.py b/pym/_emerge/depgraph.py
> index 04e724d..8a614c4 100644
> --- a/pym/_emerge/depgraph.py
> +++ b/pym/_emerge/depgraph.py
> @@ -6649,6 +6649,21 @@ class depgraph(object):
>   # will be appropriately reported as
> a slot collision # (possibly solvable via backtracking).
>   pkg = matches[-1] # highest match
> +
> + if
> (self._dynamic_config._allow_backtracking and
> + not
> self._want_installed_pkg(pkg) and (dep.atom.soname or (
> + dep.atom.package and
> dep.atom.slot_operator_built))):
> + # If pkg was already
> scheduled for rebuild by the previous
> + # calculation, then pulling
> in the installed instance will
> + # trigger a slot conflict
> that may go unsolved. Therefore,
> + # trigger a rebuild of the
> parent if appropriate.
> + dep.child = pkg
> + new_dep =
> self._slot_operator_update_probe(dep)
> + if new_dep is not None:
> +
> self._slot_operator_update_backtrack(
> + dep,
> new_dep=new_dep)
> + continue
> +
>   if not self._add_pkg(pkg, dep):
>   return 0
>   if not
> self._create_graph(allow_unsatisfied=True): diff --git
> a/pym/portage/tests/resolver/test_slot_operator_complete_graph.py
> b/pym/portage/tests/resolver/test_slot_operator_complete_graph.py new
> file mode 100644 index 000..1d59bce --- /dev/null
> +++ b/pym/portage/tests/resolver/test_slot_operator_complete_graph.py
> @@ -0,0 +1,141 @@
> +# Copyright 2017 Gentoo Foundation
> +# Distributed under the terms of the GNU General Public License v2
> +
> +from portage.tests import TestCase
> +from portage.tests.resolver.ResolverPlayground import (
> + ResolverPlayground,
> + ResolverPlaygroundTestCase,
> +)
> +
> +class SlotOperatorCompleteGraphTestCase(TestCase):
> +
> + def testSlotOperatorCompleteGraph(self):
> +
> + ebuilds = {
> + "app-misc/meta-pkg-2" : {
> + "EAPI": "6",
> + "DEPEND": "=app-misc/B-2
> =app-misc/C-1  =app-misc/D-1 =dev-libs/foo-2",
> + "RDEPEND": "=app-misc/B-2
> =app-misc/C-1 =app-misc/D-1 =dev-libs/foo-2",
> + },
> +
> + "app-misc/meta-pkg-1" : {
> + "EAPI": "6",
> + "DEPEND": "=app-misc/B-1
> =app-misc/C-1  =app-misc/D-1 =dev-libs/foo-1",
> + "RDEPEND": "=app-misc/B-1
> =app-misc/C-1 =app-misc/D-1 =dev-libs/foo-1",
> + },
> +
> + "app-misc/B-1" : {
> + "EAPI": "6",
> + "DEPEND": "dev-libs/foo:=",
> + "RDEPEND": "dev-libs/foo:=",
> + },
> +
> + "app-misc/B-2" : {
> + "EAPI": "6",
> + "DEPEND": "dev-libs/foo:=",
> + "RDEPEND": "dev-libs/foo:=",
> + },
> +
> + "app-misc/C-1" : {
> + "EAPI": "6",
> + "DEPEND": "dev-libs/foo:=
> app-misc/B",
> + "RDEPEND": "dev-libs/foo:=
> app-misc/B",
> + },
> +
> + "app-misc/C-2" : {
> + "EAPI": "6",
> + "DEPEND": "dev-libs/foo:=
> app-misc/B",
> + "RDEPEND": "dev-libs/foo:=
> app-misc/B",
> + },
> +
> + "app-misc/D-1" : {
> + "EAPI": "6",
> + "DEPEND": "dev-libs/foo:=",
> + "RDEPEND": "dev-libs/foo:=",
> + },
> +
> + "app-misc/D-2" : {
> +  

Re: [gentoo-portage-dev] [PATCH] depgraph: trigger slot operator rebuilds via _complete_graph (bug 614390)

[Zac Medico]

On Fri, Mar 31, 2017 at 10:06 PM, Brian Dolbec  wrote:
> On Fri, 31 Mar 2017 21:11:41 -0700
> Zac Medico  wrote:
>
>> Fix _complete_graph to trigger rebuilds of parent packages when they
>> pull in installed packages that had already been scheduled for rebuild
>> by the previous calculation.
>>
>> X-Gentoo-bug: 614390
>> X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=614390
>> ---
>>  pym/_emerge/depgraph.py|  15 +++
>>  .../resolver/test_slot_operator_complete_graph.py  | 141
>> + 2 files changed, 156 insertions(+)
>>  create mode 100644
>> pym/portage/tests/resolver/test_slot_operator_complete_graph.py
>>
>
> looks good.  I look forward to getting less conflicts with this...

Thanks, pushed:

https://gitweb.gentoo.org/proj/portage.git/commit/?id=a83bb83909c5a6ac232c8eb5931b28027f4175af

-- 
Thanks,
Zac